Content uploaded by Peter F. Hess
Author content
All content in this area was uploaded by Peter F. Hess on Oct 09, 2024
Content may be subject to copyright.
See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/382424558
Understanding the Crucial Role of Temporal Factors In Western Lake Erie Basin
Wetlands: When Do Wetlands Act as Nutrient Sinks or Sources?
Preprint · July 2024
DOI: 10.13140/RG.2.2.31849.30567
CITATIONS
0
READS
33
1 author:
Peter F. Hess
31 PUBLICATIONS1 CITATION
SEE PROFILE
All content following this page was uploaded by Peter F. Hess on 20 July 2024.
The user has requested enhancement of the downloaded file.
1
Understanding the Crucial Role of Temporal Factors In Western
Lake Erie Basin Wetlands: When Do Wetlands Act as Nutrient
Sinks or Sources?
Peter F. Hess P.E., BCEE, QEP
July 20, 2024
Abstract
Determining when a wetland functions as a sink or source of nutrients, especially for Dissolved
Reactive Phosphorus (DRP), is a crucial question about the viability of artificial wetlands to
reduce cyanobacterial harmful algal blooms (cyanoHABs). Most wetlands are assumed to
function as both a sink and a source of nutrients. However, it is of utmost importance not to
construct a wetland that converts from a sink to a source of DRP during the 'spring loading
period' from March 1st through July 31st. This period is known as the critical cyanoHAB
precursor loading period. Published research provides evidence that most cyanoHABs forming
DRP are delivered to Lake Erie tributaries 10% of the time when high-water flow (discharges)
occurs. The Ohio Domestic Action Plan (DAP) calls for a 40 percent total spring load reduction
in total and dissolved reactive phosphorus (TP and DRP) entering Lake Erie’s western basin. The
recently federally approved OhioEPA 2023 Maumee River Watershed Nutrient TMDL (TMDL)
relies on over 92 metric tons per year of phosphorus reductions during the spring loading from
constructing wetland and stream buffer projects managed by the Ohio Department of Natural
Resources (ODNR) at an estimated cost of $33 million. To meet the cyanoHABs mitigation
target, the creation, restoration, and enhancement of instream wetlands of the Maumee River
Basin must be designed and function as sinks rather than sources during high-water discharge
and high-water nutrient loading scenarios, especially during the spring loading period. This
analysis underscores the weight of our decisions in careful wetland construction and
management, as any oversight could significantly affect mitigation efforts.
I. Overview & Background of the Science of a Wetland Functioning as a
Sink and a Source of Nitrogen and Phosphorus Nutrients
Wetlands should be viewed as a solution and a collective effort towards a better watershed.
Evaluating the benefits of the creation, reconstruction, and enhancement of wetlands should not
solely focus on its ability to improve water quality but also consider the ancillary benefits it
provides, such as flood control, wildlife habitat, and recreational opportunities. The potential
impact of these wetland projects on the Western Lake Erie Basin is substantial, underscoring the
importance of our collective efforts and shared responsibility in this endeavor. This realization
should instill a sense of purpose and urgency in our actions.
Hydrology and biochemistry, spurred by water discharge and nutrient loading, are critical factors
in determining when a wetland functions as a sink or a source of cyanoHABs. The following
figure illustrates the pathways of nutrient mobility in a wetland.
2
Figure A. Visual depiction of how phosphorus moves within a wetland. Arrows correspond to
pathways by which phosphorus is retained (white arrows) or released (black arrows). Retention
pathways include settling, sorption, and uptake into plants and microorganisms; release pathways
include decomposition of organic material, desorption, and resuspension by flow or bioturbation.
The inserted panel enumerates factors affecting the relative strength of each pathway; for
example, low flow levels promote settling, while high flow levels promote resuspension. (Ury et
al., 2023)
A wetland associated with a waterway functions as a sink and a source of nutrients (P and N)
during various times of the hydrological year. This review will focus primarily on instream
(flow-through) wetlands with some applicability extending to depressional, floodplain, and
coastal wetlands. The following discussion will, using recent publications, explore the theoretical
and practical science surrounding when and under what circumstances a wetland performs as a
sink or a source of nutrients. (H2Ohio Wetland Monitoring Program 2021. H2Ohio Wetland
Monitoring Program Plan. Kinsman-Costello et al. provide an expansive delineation of wetland
overview and background, wetland and water quality background, nutrient function, and other
pertinent information).
This review will be structured around three significant points, providing a clear roadmap for the
readers to follow and ensuring they are well-informed. These points will guide the reader
through the key aspects of when a wetland functions as a sink or a source of nutrients,
particularly for Dissolved Reactive Phosphorus (DRP).
1. Settling suspended sediment containing nutrient DRP from a waterway to a wetland.
The conditions associated with the sediment being retained within the wetland and
when nutrient DRP is released from the sediment and reenters the waterway to
participate in the formation of cyanoHABs.
2. The “cycling” of phosphorus and nitrogen nutrients into and from suspended
sediment in a wetland and waterway through the biochemical processes.
3
3. When the DRP retention threshold in a wetland is exceeded and whether that time
coincides with the critical nutrient loading period of March 1st through July 31st when
nutrient releases contribute to cyanoHAB formation.
Whether a wetland functions as a sink or a source depends on whether the wetland retains or
releases nutrients. Under varying conditions, a wetland is a nutrient collector (sink), sometimes
releasing the previously retained nutrient (source). Biochemical processes alter the magnitude
and bioavailability of phosphorus in determining whether a wetland is a sink or a source of
cyanoHABs precursors in the Maumee River and the Western Lake Erie Basin (WLEB)
Watersheds. The critical loading period for the formation of cyanoHAB is between March 1st and
July 31st; the release of phosphorus and nitrogen nutrients (especially DRP) during this period
will interfere with attaining the phosphorus TMDL and Annex 4 targets.
The uniqueness of the soil characteristics in the Western Lake Erie basin, especially the Maumee
River watershed, is a significant factor in analyzing whether a wetland functions as a sink or a
source and the instream processing of DRP. As illuminated in the analysis conducted by the
OhioEPA, there are differences between the soil characteristics (Iron-Aluminum Oxides,
Calcium Carbonate) in the watersheds (St. Joseph River, Tiffin River, St. Marys River, Upper
Maumee River, Lower Maumee River, Auglaize River, Blanchard River) north and south of the
Maumee River, and between the watersheds (Maumee River, Portage, Sandusky) in the western
Lake Erie basin as shown by the estimated hydrological weighting factor (DAP 2023, Tables
A12-A18). The large-scale prevalence of the incorporation of crop drainage tiles in the Maumee
River basin significantly impacts hydrology, water discharge, and nutrient transport. This
Maumee River basin drainage tile placement does not homogenize streamflow metrics between
high and low discharge conditions as in Iowa. Studies in Iowa indicate that tile drainage affects
the baseline portion of the hydrograph. Contrasting results in Ohio show that because tile drains
are placed shallower in the soil, there is a more direct connection between precipitation and
nutrient concentrations, loads to the recipient streams and rivers where the instream wetlands are
located. The above demonstrates that watersheds with the most significant percentage of tile
drainage have more chemostatic concentration versus discharge behavior (Miller and Lyon,
2021).
Characterizing the hydrodynamic science of sediment in a water column has been studied for
many years and is the subject of many textbook chapters. These textbooks focus on the
simulation of settling organic particles produced by or causing eutrophication, their
decomposition, or release from the bottom sediment or the water column. Although the subject is
extraordinarily complex, using Stoke’s Law, fluid mechanics, and particle dynamic theorems can
create suspension, deposition, and resuspension models in a distributed system (Chapra, 2008;
Droste and Gehr, 2019).
Suspended solids containing nutrients are transported and transformed by many different
mechanisms. The determination of whether a wetland functions as a sink or functions as a source
is dependent upon the following significant hydrological factors:
A. The concentration and composition of the nutrients can be adsorbed by suspended
solids (sorption-desorption equilibrium). The amount of P and N entering the
4
watershed at any time (flux), the legacy amount of P, suspended DRP (SRP), and N
contained in the bottom sediment (sediment nutrient saturation and sediment
physiochemical characteristics). Whether the nutrients have a high concentration of
Total Phosphorus (TP) versus DRP, which is 100% bioavailable, and the portion of
the DRP that is colloidal-P, which may be less bioavailable.
B. The velocity of the water flowing through the wetlands. The velocity of the water in a
ditch, creek, stream, or river (prior to and following the watershed) determines the
sediment's settling velocity and residence time and whether it contains nutrients.
C. The water produces turbulent energy before moving into, following, or while in the
wetland.
D. Temperature and water depth.
E. The abundance of vegetation and the removal of decayed organic matter.
F. The particle diameter and the composition of the suspended solids.
The manner of wetland construction and its location play a significant role in determining when
they coincide with higher hydraulic loading rates, lower influent P concentrations, and legacy
soil/sediment phosphorus. While wetlands are generally effective at removing N
(49.4 ± 25.4% removal efficiency ). Wetlands are complex ecosystems, with their efficacy for
retaining P being more variable (Land et al., 2016) and even a source of P for downstream water
(Audet et al., 2020). Seasonal patterns and changes to the sorption-desorption equilibrium of P
retention are more important for understanding wetlands' role in P-driven water quality
problems. Wetlands switch between functioning as sources and sinks multiple times yearly (Ury
et al., 2023). However, extreme rain events are one of the leading causes of poor P retention,
accounting for significant portions of annual wetland P export (Ardon et al., 2010; Jiang and
Mitsch, 2020). Managing these events is crucial for effective wetland nutrient dynamics. The
alternation between dry and wet cycles within a wetland has been shown to promote the
metabolism of organic matter and the release of organic P (Kieckbusch and Schrautzer, 2007).
As discussed above, the degree of N and P nutrient removal varies within a wetland functioning
as a source or sink. The crucial point is that a wetland reduces both nutrients. This N removal or
contribution will impact the toxicity of the cyanoHABs that produce many toxic secondary
metabolites called cyanotoxins. The most studied group of cyanotoxins are microcystins (MC),
with over 300 congeners reported. MC-LR is the most studied congener because of its abundance
and toxicity. Recent toxicology studies suggest that more hydrophobic MC congeners such as
MC-LA, MC-LF, and MC-LW may be less abundant but up to seven times more toxic than MC-
LR. In contrast, MC-RR’s toxicity is only one-fifth that of MC-LR. Hence, understanding the
environmental stressors that change the MC congener profile is critical to assessing the negative
impact on environmental and human health. A two-year field and experimental study
investigated seasonal and spatial changes of MC congener profiles in the western basin of Lake
Erie. Both studies showed that nitrogen enrichment favored the production of nitrogen-rich MC-
RR (C49H75N13O12). The field study showed that nitrogen depletion favored the low-nitrogen
MC-LA (C46H67N7O12). MC-LR (a medium N level, C49H75N10O12) accounted for ~30% to 50%
of the total MC concentration and was stable across nitrogen concentrations. Using each MC
congener's relative toxicity and concentrations, relevant analytical analysis (enzyme-linked
immunosorbent assay and liquid chromatography-mass spectrometry analysis) overestimated the
toxicity in early bloom (July) and underestimated it in late bloom (September). On 24 July 2019,
5
highly toxic MC-LW and MC-LF were detected at nearshore stations with relative toxicity
exceeding drinking water standards. This study demonstrated that the less toxic, high-nitrogen
MC-RR dominated under nitrogen-replete conditions in the early season. In contrast, the more
toxic, less nitrogen MC-LA dominated under nitrogen-limited conditions later in the season
(Chaffin et al., 2023).
Figure B. Conceptual diagram illustrating wetlands' dominant nitrogen and phosphorus removal
mechanisms. Wetlands can remove P by three primary mechanisms: plant uptake and tissue
storage, geochemical association of phosphate with sediment minerals (e.g., sorption to iron
oxides), and long-term storage by sedimentation and burial. None of these processes permanently
remove P from an ecosystem, and stored P may be re-released and exported downstream as plant
tissues senesce and mineralize, associated phosphate is released from sediment minerals, and
particulate P is resuspended and transported in surface waters. Nitrogen most commonly enters
wetland systems as nitrate (NO3) from surface runoff. In addition to removal via plant uptake,
settling, and burial, the wetland removal pathway exports N to the atmosphere. However,
incomplete denitrification and turnover of particulate N can release readily available forms of N
(e.g., urea, NH4+) that can be exported to downstream systems (H2Ohio Wetland Monitoring
Program 2021. H2Ohio Wetland Monitoring Program Plan. Kinsman-Costello et al., page 10).
The knowledge of sediment-water interactions is critical to determining the capability of a
wetland to be a nutrient sink. Bottom sediments have long been acknowledged as a potential
source of P (DRP, SRP, and non-soluble P) to the overlaying waters of lakes and impoundments.
As discussed in the Monitoring Plan, semi-empirical formulations can simulate sediment nutrient
feedback and retention. A simple season approach will not deliver a realistic view of the
interactions necessary to illuminate the operational characteristics of a wetland's sediments,
hypolimnion, and epilimnion segments. A multi-seasonal evaluation focusing on the March-July
period, considering the nutrient (DRP & N) wetland loading of other seasons, is a preferred path
for a diligent evaluation.
6
Consideration of incorporating into the modeling the nitrogen cycle (ammonia and nitrate
assimilation, ammonification, nitrification, denitrification, and nitrogen fixation) in conjunction
with the phosphorus cycle as part of the nutrient production/decomposition life cycle will
evaluate the toxicity of microcystins. Due to the fluctuation of water discharge, the increase of
heat balance, the variability of nutrient flux from the surroundings, and changing wind patterns,
the development of unique integrated kinetic segmented models is the only vehicle that should be
used to incorporate science with the policy deciding for the location of, and the potential
effectiveness of wetlands.
Although not a new subject, quantifying the nutrient removal (nitrogen and phosphorus
compounds) by a constructed wetland dominated by diffuse agricultural groundwater inflows has
been a subject of considerable review. There are numerous similarities between the science of
wetlands as either sources or sinks and the “Instream Processing” of nutrients discussed in
Section 4.1.4 of the 2023 Domestic Action Plan (DAP 2023, pg. 68-71) that address instream
processes such as biological activity or sedimentation which capture and release P and N
compounds, especially DRP.
The soil exchange potential of a wetland site should be used in conjunction with hydraulic and
biochemistry analysis and stream wet discharge characteristics (Stokes Law) to determine
whether a wetland functions as a sink or a source of sediment-P during the numerous seasonal
water flow cycles. It is essential to explore the P saturation of the sediment in the basins and the
impact of variables of transfer between the water column and the bottom sediment.
As mentioned earlier, the characteristics of the sediment influence the watershed reach scale
drivers of DRP release and retention. The determination of whether a wetland fits the label of a
sink or a source is influenced by the physicochemical characteristics of sediment the watershed
collects, the amount of DRP in the streams, and sediment phosphorus saturation (amongst other
factors) (Kreiling et al., 2023). A review of the hydrology of St. Joesph, St. Mary’s, Tiffin,
Auglaize, and Blanchard headwaters in various HUC12 regions in the Maumee River, Portage,
Cedar Creek, and Toussaint Watersheds indicates the smaller area watersheds have the highest
potential of wetlands being a source due to high discharge. The streamside wetland sediment
collection potential has changed between 2019 and 2021. With sediment-P concentration being
the predominant variable (sediment structure being a constant), it can be assumed that the influx
of DRP in specific regions drives the P exchange potential. The factors mentioned above should
drive the reconstruction of wetlands. A possible explanation for the change may be the recent
termination of in-lake silt disposal created by dredging Maumee Bay.
The issue of a wetland operating as a sink or a DRP source depends upon soil P saturation, P
loading, and legacy P constraints. Often, the economy of a landowner willing to allow a wetland
overshadows the efficacy of a wetland at a location (Berkowitz et al., 2020). The location
selection for a wetland is critical because the amount of loading in a basin or to the streams and
rivers feeding the wetland dictates the potential removal of sediment-P. The ability of a wetland
to remove DRP is greater than the ability of a wetland to remove TP (Mitsch, 2017). The
Maumee River Nutrient TMDL only sets a TP target without a DRP target advocated by the
Great Lakes Water Quality Agreement (GLWQA). The 2023 DAP contains DRP and TP targets
based on loading and concentration.
7
Wetlands employ three central phosphorus sorption mechanisms: a) plant uptake and tissue
storage, b) geochemical association of phosphate with sediment minerals, and c) long-term
storage through sedimentation in the retention of nutrients transported from crop fields or
through the Maumee River, Portage River, Toussaint Creek, and the Sandusky River watersheds.
Figure C illustrates the sorption of the organic compounds DRP ( PO43-) by suspended
sediments and the settling and decomposition of the same. Leaving phosphorus nutrients
absorbed onto suspended solids in the water column and bottom sediments. Sorbed organic P
may be released from sediments and soils if conditions alter sorption equilibrium (Noe et al.,
2013).
Figure C. Examination of interactions between DRP and suspended sediment-P during high flows
in the Maumee River network, focusing on March-June DRP exports, which fuel HABs in Lake
Erie. The suspended sediment influences the DRP-sediment-P interactions in the waterways. The
lower quantity of suspended sediment leads to greater DRP; high flows influence both (King et
al., 2022). Wetlands are a mixing basin/settling chamber for suspended sediment and DRP.
Figure D below illustrates the utilization of the nutrients in the growth of plant matter living
within the wetlands (nutrient uptake). Even the sorption of nutrients within a wetland is not
permanent because plants are composed of organic matter; when the plants die in the winter
and begin decomposition, the more decomposable fraction is recycled back to the water
column.
8
Figure D. Plants and microbes can directly assume phosphate, converting inorganic P to organic P
forms. Organic P is mineralized to phosphate via microbial decomposition. The phosphate ion is
geochemically reactive to many soil minerals. It can sorb strongly to iron and aluminum oxides and
co-precipitate with calcium carbonates. The strength of the sorption-desorption equilibrium with metal
oxides depends on metal redox forms, mineral structure, and soil pH conditions (H2Ohio Wetland
Monitoring Program 2021. H2Ohio Wetland Monitoring Program Plan. Kinsman-Costello et al., page
11).
In wetlands, as well as in streams and rivers, the biochemical processes alter the magnitude and
bioavailability (e.g., DRP versus sediment-P) of phosphorus transported from the wetlands to
receptor bodies during low discharges and, subsequently, the formation of cyanoHABs (Withers
and Jarvie, 2008). This phenomenon, called cycling, is due to the suspended sediment and DRP
mixing within a wetland and the sorption-desorption of DRP during high-flow turbulent
conditions (King et al., 2022). One potentially important aspect of river phosphorus cycling
during high flows is phosphorus sorption- desorption by suspended sediment, which can increase
or decrease DRP concentrations and, therefore, the bioavailability of phosphorus exports to
cyanoHABs (King et al., 2022)(Pennuto et al., 2014). The cycling of DRP versus sediment-P is
part of the legacy phosphorus and the inability to establish a DRP target for the OhioEPA 2023
Maumee River Basin Nutrient TMDL.
A water stream contains both P and N nutrients, and a wetland processes the nutrient contents of
the water stream. The H2Ohio Wetland Monitoring Program:2021 correctly includes N
compounds in its studies because its constituents play a significant role in developing toxic
cyanoHABs in Lake Erie. “A recent review suggests that management efforts to reduce P
pollution without controlling N have caused nutrient imbalances in eutrophic systems, which
may favor toxic cyanobacterial HABs (microcystins) that cannot fix atmospheric N2 gas.
Controlling P has received much policy focus; farmers have added on average, twice the amount
of commercial N fertilizer (15.9% urea, 32% urea-ammonium-nitrate, 37.5% ammonia, and
9
14.7% other) as P fertilizer per acre over the last few years (Prokup et al., 2017), exacerbating
the eutrophication problem. This lack of understanding of the N cycle due to a greater focus on P
may indicate that human actions meant to help solve the problem (e.g., the US and Canada
committed to a 40% reduction in P loading to Lake Erie; IJC 2016, Annex IV) may have
exacerbated HABs in the western basin of Lake Erie by promoting slightly smaller but more
toxic blooms” (Gobler et al., 2016, and H2Ohio Wetland Monitoring Program 2021. H2Ohio
Wetland Monitoring Program Plan. Kinsman-Costello et al., pages 13 and 14)
+
Nitrification
Figure E. The nitrogen (N) cycle is characterized by numerous microbially mediated redox
transformations of N among its many chemical forms and oxidation states. From the water quality
perspective, the optimal N removal mechanism is complete denitrification, in which nitrate is
converted to inert nitrogen gas (N2). Yellow arrows indicate oxidation reactions, red arrows are
reduction reactions, and white arrows indicate no redox change (H2Ohio Wetland Monitoring
Program 2021. H2Ohio Wetland Monitoring Program Plan. Kinsman-Costello et al., page 13).
The location characteristics, input nutrients, and many other factors influence the sink versus
source criteria for flow-through, coastal, floodplain, and depressional wetland ecosystems. The
temporal loading characteristics exhibited in the Western Lake Erie watersheds of the Maumee
River, Sandusky River, and Portage River may be most important. The uniqueness of the
Maumee River basins' temporal loading characteristics counters old nutrient assimilation and
nutrient release assumptions (Risgaard-Petersen and Ottosen, 2000).
Figure F illustrates that the “fill-and-spill” or “hockey-stick model” Bayesian Changepoint
calculations demonstrate temporal assimilation and release assumptions. The classification of a
wetland as “sometimes a sink and sometimes a source” must be replaced by whether the wetland
functions as a “sink or a source of nutrients during the March 1st through July 31st period.”
10
Therefore, the ability of wetlands to sequester nutrients during the critical spring months is more
important in reducing HABs than the net annual nutrient sequestration (H2Ohio Wetland
Monitoring Program 2021. H2Ohio Wetland Monitoring Program Plan. Kinsman-Costello et al.,
page 25.).
The hockey-puck modeling in Figure F of ‘Old Women’s Creek,’ a flow-through artificial
wetland, shows a sorption capacity of 3,760 pounds/year of phosphorus load in the H2Ohio
Wetland Monitoring Program Plan. Kinsman-Costello et al. on page 32. This amount is
calculated using the Bayesian Changepoint model and does not delineate the sink v. source
characteristics during high-water discharge and high-nutrient input conditions. As demonstrated
later, the high-water discharge and high-nutrient input conditions are shown through a temporal
plotting of the “capacity,” which may be a theoretical assumption unrelated to real-world
conditions. Also, Figure F provides no data about the quantity of phosphorus released from the
wetland. The larger question is whether the 3,760 pounds P/year is retained from March 1st
through July 31st.
In order to identify ‘red flags’ of the “Old Women’s Creek” wetland and all constructed
watersheds, all versions of Figure F should include the following.
• The calculated quantity of P release (right of the equilibrium line) during the same
period,
• A delineation of the sorption and desorption on a historical basis, related to March-July,
• A delineation of the water discharge period used to construct the chart,
• The chart is constructed based on high-low-mean water discharges and nutrient flux.
The use of phosphorus retention estimate based upon this calculation in determining
conformance with the reduction target contained in the TMDL should be highly suspect because
“When graphed, the data forms a “hockey stick” pattern: at low inflow TP loads, outflow TP
concentrations are relatively stable and low. Conversely, at higher inflow TP loads, outflows TP
concentration increases as TP loading increases and are more variable (H2Ohio Wetland
Monitoring Program 2021. H2Ohio Wetland Monitoring Program Plan. Kinsman-Costello et al.,
Page 31).”
11
Figure F. (A) Illustration of the “fill-and-spill” or “hockey stick” conceptual model for wetland
Total P retention. A wetland has a specific capacity to retain inflowing TP loads. For inflowing
TP loads below the TP retention threshold, TP concentrations in the wetland outflow are
relatively stable and low. When TP loads exceed a given wetland’s TP retention threshold, TP
export from the wetland increases and becomes more variable. (B) The changepoint model shows
the relationship between wetland inflow TP loading and outflow TP concentration. The point of
inflection (φ) is an estimate of the TP retention threshold. Modified from Qian and Richardson
(1997). (C) TP loads (x-axis), TP outflow concentration (y-axis) in the Old Woman Creek
wetland. The Bayesian Changepoint model results indicate that the TP retention threshold for this
system is 3,790 lbs. P per year. (H2Ohio Wetland Monitoring Program 2021. H2Ohio Wetland
Monitoring Program Plan. Kinsman-Costello et al., Page 32).
12
During the design of wetland and edge-of-field nutrient reduction strategies, the science of
hydrology, biochemistry (DRP cycling that creates a legacy phosphorus reservoir), and
engineering/science applications must be strictly followed to ensure sediment-P and DRP capture
by incorporating the correct settling velocity for high-water and medium-water discharge
scenarios. Proper engineering will determine if a wetland or edge-of-field mitigation will
function as a sink rather than a source of nutrients. Maumee River/ Western Lake Erie watershed
nutrient reduction strategies must ensure high-water discharges will prevent a wetland from
functioning as a source rather than a sink. Additionally, high-water discharges must be the focus
of engineering edge-of-field practices (e.g., bioretention cells and saturated buffers, smaller-sized
wetlands, and stream calming) to prevent high-water flow, which results in extreme discharge
events that are likely to create an ineffective nutrient sink. Even those practices built to provide
nutrient capture become inundated during high-water flows (e.g., two-stage ditches). They are
unlikely to provide effective nutrient removal due to the reduced retention and contact time
during high-water flows. For example, a two-stage ditch implemented in an agricultural stream
located in Indiana reduced annual NO3-N loads by only 2- 3% when inundated (Speir et al.,
2020). These results should motivate a shift from removing nutrients at the edge of the field (as
most practices provide poor removal during high-water discharges) to reducing the nutrient
inputs using other innovative practices (King et al., 2022). This example shows that regulated
reductions (St Mary’s Lake) may be the exception and not the rule when determining the
efficacy of wetlands and stream buffers as a sink of DRP and TP.
Nutrient cycling is shown in the recent long-term P mass balance study, which includes the P
contributions of sinks and sources (Bocaniov et al., 2023 (1); Bocaniov et al., 2023 (2)). The
multiyear study demonstrates the need for internal load reduction and articulates the load
mechanisms creating cyanoHABs in Lake Erie with the differentiation between the basins. The
mass balance shows the multitude of phosphorus contributors and removal systems. Various
inputs included in the mass balance, such as atmospheric deposition, were not incorporated in the
TMDL.
High-water discharges significantly affect the DRP transported to the Lake Erie basin. This is not
only due to a wetland's functioning as a sink or a source of DRP but also to the cycling of DRP
from bottom sediment into watersheds as a form of legacy precursors of cyanoHABs.
The use of flow-weighted mean concentration (FWMC) and flow-weighted mean load (FWML),
a tool to normalize water discharge, should be carefully weighed against the substituted TP and
DRP non-normalized load and concentration as a measurement of progress. Water discharge is
crucial in cycling bottom sediment (legacy) cyanoHAB precursors. Attainment of the TP and
DRP targets must account for legacy contaminants that are high-water discharge dependent, as
seen in Lake Erie mass balances (Bocaniov et al., 2023. (2)).
13
Figure G. Mean TP fluxes 2003 to 2016 in metric tons per annum. Loadings may not add up due
to rounding errors. The DRP mass balance of the internal load from cycling through
sedimentation DRP exchange is significant (Bocaniov et al., 2023. (2)).
Most certainly, an increase in the temperature of the lake water and the air and oscillation of the
water body due to wind action (seiches) factor into the variables of sediment-P water column
transfer in mass balance calculation. The measured change of phosphorus exchange potential is a
lake and a tributary phenomenon (Kreiling et al., 2022), as shown in Figure H. The US EPA
noted in comments to the 2023 OhioEPA draft TMDL that western Lake Erie water body
impacts due to ‘Climate Change” is a factor worthy of consideration.
14
Figure H. Phosphorus exchange potential in 2019 (a) and 2021 (b) sites. The exchange potential
is a surrogate determination of whether a wetland area may be a sink or a source of sediment -P.
Sites with negative phosphate exchange potential were potential phosphorus sites during
sampling (Kreiling et al., 2023).
One significant removal system that was not included in the Bocaniov mass balance is the impact
of the annual removal of 850,000 to one million cubic yards of nutrient-laden sediment due to
the annual dredging of the Maumee Bay and Maumee River by the US Army Corps of Engineers
15
coupled with the near-term prohibition of open-lake -placement of the dredging spoil into Lake
Erie (Hull et al., 2012; USCOE, 2023; Bhatia et al., 2014). Dredging and the environmentally
sound land disposal of the silt federal shipping channel (silt contains 10-12% bioavailable P in
the TP (Toledo Lucas County Port Authority, 2020)) impacts the recycling of legacy P and
cyanoHAB formation. The cycling of DRP and its release under reducing conditions (Gibbons
and Bridgeman, 2020; Matisoff et al., 2016; Parsons et al., 2017) is a vital issue to be explored.
The mass balance study calculates the watershed inputs to be 3,020 metric annual tons,
sedimentation removal at 3,188 metric tons (not including silt dredging), and atmospheric
deposition at 127 metric tons. The release of legacy DRP in the bottom sediments is a subtle
positive source, but the main driver of cyanoHABs is the external load.
The following points may be ascertained from the review of the science of whether wetlands are
a sink and a source of DRP in the formation of cyanoHABs.
• The science of settling velocity (Stokes Law), hydrology, biochemistry, and associated
studies indicate that a high-water velocity will generate a higher discharge of bioavailable
DRP and sediment from a wetland, streambed, and river than a low water velocity and
discharge.
• Wetlands (and numerous edge-of-stream measures such as bioretention cells, saturated
buffers, and two-stage ditches) do not function as a sink during high water discharge and
subsequently release bioavailable sediment-DRP. They capture sediment-DRP during
low-water discharge. Multiple biochemical and hydraulic factors influence the prevention
and release of DRP from wetlands, which causes the formation of cyanoHABs
downstream.
• Sediment-DRP interactions (cycling) demonstrate that DRP is transported long distances
during high-water discharges and is a factor in the formation of cyanoHAB.
• Wetlands may not reduce cycling during high-water flow, which is a factor in forming
legacy phosphorus in Lake Erie and cyanoHAB.
• N, in conjunction with P, contributes to cyanoHAB formation and warrants reduction in a
wetland.
II. High-Water Discharge Corresponds with High Nutrient Loading in the
Western Lake Erie Basin Converting Wetlands from Sinks into Sources
The following information will briefly review the studies demonstrating that high-water
discharge rates from wetlands function as DRP sources rather than as DRP sinks.
A. Wetlands and other conservation practices function as a source of DRP-suspended sediment
rather than functioning as a sink during high-water discharge conditions.
B. The top 10% of the highest water discharge conditions transport the majority of DRP
cyanoHAB forming nutrients from the Maumee River, Portage River, Sandusky, and
Toussant Creek watersheds to Western Lake Erie Basin according to streamflow curves.
• 54% from the Maumee River watershed,
• 63% from the Blanchard watershed,
• 44% from the Tiffin watershed,
16
• 74% from the Portage watershed,
• 64% from the Sandusky watershed.
C. The top 40% of the highest water discharge conditions transported most of the nutrient load
from the Western Lake Erie Basin to Lake Erie according to streamflow curves.
• 55-94% of the annual dissolved reactive phosphorus (DRP) loads,
• 54-98% of the annual nitrate and nitrite (NO3-N) loads,
• 79-99% of the annual total phosphorus (TP) loads,
• 86-99% of the annual total suspended solid (TSS) loads.
The H2Ohio nutrient reduction strategies focus on implementing best management practices
(BMPs) or agricultural conservation practices (ACPs), which include nutrient management
practices (e.g., 4R management- right source, right rate, right time, right place), land use changes
(e.g., land retirement and grazed pasture), infield practices, (e.g., conservation tillage and cover
crops, and the edge of field practices, (e.g., wetlands, riparian buffers, lined inlets, bioreactors, or
two-stage ditches) (Baker et al., 2018) (Christianson et al., 2018). Studies conducted in
agricultural wetlands found that a majority of the nutrient loads, NO3-N, DRP, and TP, were
released in the upper 10% of the discharges(Q > Q10%) for the periods investigated. The critical
timing of wetland nutrient release is as important as implementing the 4-R strategy for BMPs
and ACPs.
The key facts based on the science of a wetland functioning as a sink or as a source of sediment-
P can be summarized as follows.
• The concentration and composition of the nutrients that suspended solids can adsorb.
The amount of phosphorus entering the watershed at any time (flux), the legacy
amount of phosphorus suspended or as bottom sediment. Whether the nutrients have a
high concentration of TP versus DRP, which is 100% bioavailable, and the portion of
the DRP that is colloidal-P, which may be less bioavailable.
• The velocity of the water flowing through the wetlands. The velocity of the water in a
ditch, creek, stream, or river (prior to and following the watershed) determines the
settling velocity of the sediment, whether it contains nutrients,
• Turbulent energy produced by the water prior to moving into, following, or while in
the wetland,
• Temperature and water depth,
• The abundance and species of vegetation and the removal of decay of organic matter,
• The particle diameter and the composition of the suspended solids.
The subsequent correlation is the measurement of water discharge rates at the monitoring gauges
in the watersheds feeding the Western Lake Erie Basin as they relate to nutrient flux. Numerous
recently published studies have measured and evaluated streamflow. Table A and Figure I
provide a graphical summary of streamflow curves and nutrient flux.
Streamflow curves were divided into five hydraulic intervals: High-water flow (0-10th
percentile), Moist conditions (10-40th percentile), Mid-range water flow (40-60th percentile),
Dry conditions (60-90th percentile), Low-water flow (90-100th percentile) (USEPA) and were
classified by USGS stream gauges. The measured nutrient and sediment load comparison to
17
streamflow shows when most pollutants are released (Kamrath et al., 2023). In general, the
nutrient loads were predominantly released during high-water discharge conditions. High-water
discharges are when wetlands and in-stream mitigation systems become a source of sediment-P
rather than a sink.
Table A. Summary of the high-water flow (H), moist condition (M), mid-range water flow (M/R),
dry conditions (D), and low-water flow (L) pollutant loads for nitrate (NO3-N), dissolved reactive
phosphorus (DRP), total phosphorus (TP), suspended solids (TSS). The values presented are
median annual values. The table is color-coded to highlight values as follows: dark blue with
white text >50%, medium blue between 20 and 50%, light blue between 10 and 20%,
exceptionally light blue between 5 and 10%, no color between 0 and 5%. (Kamrath et al. 2023).
Table B. Guide and information related to Table A (Kamrath et al., 2023).
High-frequency monitoring in the Little Auglaize River confirms that high-water discharge flow
events fundamentally drove nutrient loads. These events propagate the transfer of DRP from tiled
fields to the sub-waterways (Pace et al., 2022). This dominant downstream discharge is
confirmed in all Western Lake Erie basins (Miller et al., 2021).
18
Figure I. A ridgeline chart to visualize the distribution of high-water discharge load for all sites
and constituents investigated. A vertical line was created at the high-water discharge designation
m10%= 50%. Shading to the right indicated that most of the annual contaminant load was released
during high water flow (Kamrath et al., 2023).
The results show the sediment-P DRP loads during high-water flow levels in Table A, and Figure
I provides data about wetlands converting to sinks (and other streamside mitigation strategies) to
sources in the Maumee River, Sandusky River, Blanchard River, and smaller feeders into the
Lake Erie Basin. The results show a substantial difference in the nutrient load released between
the upper 40% of flows and the bottom 60% of flows (Table A). The table and figure provide a
better understanding of the relationship between streamflow and nutrient loads using high-
frequency water quality monitoring data, and the information focuses on temporal analysis.
The nutrient loads were predominately released during high-water discharge flows (Table A and
Figure I). For most watersheds, high loads (m10%) often produced greater than 50% of the annual
load, especially for TP and TSS (Figure I). This indicates that the wetland and streamside
mitigation measures were ineffective during the highest loading period of the year. This is
especially true when the high TSS loading is correlated with sediment-P. Any disparity in (m10%)
between small (Sandusky and Portage) and large (Maumee) watersheds highlights the influence
of watershed areas on the relationship between nutrient export and high-water discharge
conditions. Although the larger watersheds typically had lower (m10%) values than smaller
watersheds, their (m10%) values still represented a substantial amount of the annual load,
especially for TP and TSS. These large versus small watersheds are more likely applicable to the
sub-watersheds and headwaters of the Maumee.
During high-water discharges, the magnitude of P sorption-desorption is likely determined by
DRP concentrations, suspended sediment concentrations (TSS), and river discharge
characteristics. The influence of DRP sorption- desorption during high-water flows influences
DRP exports and, later, cyanoHAB formation.
19
During the design of wetland and edge-of-field nutrient reduction strategies, the science of
hydrology, biochemistry (DRP cycling that creates a legacy phosphorus reservoir), and
engineering/science applications must be strictly followed to ensure sediment-DRP capture by
providing an appropriate settling velocity during high-water and medium-water discharge
scenarios if a wetland or edge of field mitigation will function as a sink rather than a source of
nutrients. Maumee River/ Western Lake Erie watershed wetland nutrient reduction strategies
must target high-water discharges to prevent wetlands from functioning as sources rather than
sinks because high-water discharges must be targeted, edge of field practices, which bypass high
flows resulting in extreme events (e.g., bioretention cells and saturated buffers, smaller sized
wetlands, and stream calming) because they are likely to perform as ineffective nutrient sinks.
Even those practices built to provide treatment, which becomes inundated during high-water
discharges (e.g., two-stage ditches), are unlikely to provide effective nutrient removal due to the
severely reduced retention and contact time during high-water discharges. For example, a two-
stage ditch implemented in an agricultural stream located in Indiana reduced annual NO3-N loads
by only 2- 3% when inundated (Speir et al., 2020). These results should motivate a shift from
removing nutrients at the edge of the field (as most practices provide poor removal during high
flows) to reducing the nutrient inputs using other innovative practices (King et al., 2022).
Overall, m10%, m10-40%, and m40% values show that high-water discharges have consistently
carried most of the nutrient loads to downstream water bodies. The positive correlation between
high-water discharge load and agricultural land use within a watershed makes a case for
agricultural nonpoint source pollution during high-water flows to be the core driver of nutrient
release, especially P across the western Lake Erie basin. Effective nutrient reduction strategies
should target agricultural nonpoint source nutrient load released during high flow and moist
conditions. Furthermore, these consistent high-water flow loads suggest that edge-of-field
practices (e.g., wetlands, buffer strips, blind inlets, bioreactors, two-stage ditches) implemented
over decades might not have been effective in capturing nutrient loads during high floods from
significant precipitation events (Kamrath et al., 2023).
The expected efficiency of nutrient removal is critical to attaining water quality goals because
bioavailable P exports are typically assumed to be 8% of sediment-P and 100% of DRP loads to
Lake Erie. On average, 85% and 93% of Maumee River March- July DRP and sediment bound-P
exports occur during high flows (Table A and Figure I). The majority of suspended sediment and
presumably sediment-bound P, exported from the Maumee watershed is deposited in or near the
Maumee River mouth, which has lower rates of internal P recycling (<7% of cyanobacteria P
demand) due to the higher oxygen concentrations in overlying waters (Matisoff et al., 2016). In
reality, Maumee Bay functions as a wetland sink that is an incubator of cyanoHABs.
Nevertheless, the best way to manage cyanoHABs is to reduce P losses at the source by limiting
P application to the fields. Furthermore, with changing environmental conditions (more
significant rainfall and snow), P-bound Lake Erie sediment could fuel internal loading and
cyanoHABs (Gibbons and Bridgeman, 2020) (King et al., 2022). Characterization of the organic
and inorganic P in the sediment is essential to ascertain the magnitude of sediment-P cycling,
with the 0-10 cm portion totaling 172 metric tons P/km2 potential internal P loading in the
western basin of Lake Erie, obviously becoming a major source category. Seiche movements'
20
impact in Lake Erie must be reviewed due to the central basin's 359 metric tons P/Year available
to the western basin, where it may participate in forming cyanoHABs (Wang et al., 2021).
The following points should be ascertained from the review of the measurement of high-water
streamflow rates and nutrient concentrations at various Maumee River watersheds, Portage
River, Toussaint Creek, and Sandusky River watershed locations.
• Stream measurement data shows that for the Maumee River, 93% of the DRP, 94% of
TP, 91% of Nitrates/Nitrites, and 100% of Total Suspended Solids (including sediment-
P) loads are released to Lake Erie during high-water discharge rates that occur 40% of the
time.
• Stream measurement data shows that for the Maumee River, 54% of the DRP, 60% of
TP, 49% of Nitrates/Nitrites, and 68% of Total Suspended Solids (including sediment-P)
loads are released to Lake Erie during high-water discharge rates that occur 10% of the
time.
• During high-water discharges, wetlands become ineffective at capturing nutrients, with
85% and 93% of the DRP, sediment-P, and TP reaching Maumee Bay.
• Bayesian Changepoint modeling should be performed for other watersheds and sub-
watersheds to determine the nutrient discharge from March 1st through July 31st.
• Scenarios replicating high-water flow (H), moist condition (M), mid-range water flow
(M/R), dry conditions (D), and low water flow (L) conditions should be developed and
modeled.
• Pollutant loads produced during high-water discharges should be a target of nutrient
reduction strategies.
III. Review of Wetlands Functioning as Sources Versus Sinks of Nutrients
during the March 1st through July 31st Impacting the H2Ohio Instream
Water Management Program
The following is a review of the ODNR wetland restoration intended to remove nutrients to meet
the targets in the TMDL during the March 1st through July 31st critical loading timeframe. The
review will compare the location of existing and proposed wetlands to the high-water
measurements and the science of a wetland functioning as a sink or source of nutrients. It is
essential to understand that not all constructed wetlands will contain a water improvement
function.
The review will address the following items.
A. The ODNR wetland construction and renovation program. Incorporating 59 wetlands
(Figure J) under the water management program.
B. Applying Stokes Law hydrodynamic modeling, the 36 (28 Maumee River watershed-8
western Lake Erie watershed) inland wetlands (Figure J) are susceptible to high
streamflow turbulence (also due to high-wind for coastal wetlands) entrainment, sediment
21
resuspension, and recycling of bottom organic matter within a distributed system, turning
a wetland into a nutrient source, rather than a nutrient sink due to the flowthrough water
velocity exceeding the suspended sediment settling velocity during high flows (m10%>
s), (m10-40%>vs),(vb≪vr), wetland channeling, settling velocities of silt/clays, organic
particles, phytoplankton.
C. Coastal wetlands function as sources of DRP load if seiches occur from March 1st
through July 31st due to the bathtub effect.
D. The ODNR water management plan cannot reduce the projected phosphorus load to all
the Lake Erie tributaries to 119,000 pounds per year (DAP, pg. 24).
E. The ODNR water management program requires a realistic assessment of whether it will
attain the 92 metric ton per year reduction target for the Maumee River or the targets on
the Portage River, Toussaint Creek, and Sandusky Rivers. (OhioEPA Maumee River
Watershed Nutrient TMDL, 2023 and Table C)
ODNR developed a robust water management program to create wetlands and other capture
mechanisms in Ohio’s waterways to reduce nutrient transport to Lake Erie. The H2Ohio program
funds these water management actions.
In the multifaceted ODNR water management program, most of the completed projects are
coastal projects that, unless studied, will be unable to demonstrate a reduction of nutrients from
tributaries feeding Lake Erie. Wetland construction and rehabilitation must have a
preconstruction evaluation review component and be engineered as a water quality improvement
strategy. The five categories of wetland construction are listed below.
• In-water flow-through Coastal Wetlands: Flow-through coastal wetland projects mainly
within the Maumee and Sandusky river mouths at locations that maximize nutrient
reduction benefits and improve fish and wildlife habitat.
• Reconnecting Diked Wetlands: Projects to divert surface water flow from upland areas
into diked wetlands to process sediments and nutrients before reaching the lake.
• Nature-Based Shoreline Wetlands: Nature-based shoreline projects to control erosion and
improve near-shore water quality by filtering water flowing from small tributaries and
drainage channels into Lake Erie.
• Nutrient Processing Wetlands and Surface Water Treatment Trains: Projects such as
Grand Lake St. Mary’s treat nutrient-laden water from agricultural and urban lands.
Incoming water is captured upstream, pumped through the water control structure, and
released into engineered riparian or coastal wetlands designed to provide settlement and
nutrient reduction benefits.
• Stream Buffers, Riparian Restoration, and In-field Wetlands: Projects incorporating
vegetative and forested buffers and enhanced and restored riparian or in-field wetlands.
These projects will be located within high phosphorus load areas in the Maumee and
Sandusky River basins and combined with best management practices (BMPs) to attain
desired phosphorus reduction and water quality benefits. Wetland Project details are in
DAP 2023, ppg. 67 and 68.
22
Figure J. Map delineating Lake Erie wetland projects' is a crucial visual aid. It features dots that
indicate the approximate location of projects, with each project noted within its respective
watershed. The 2008 Annual Load TP within the Maumee River is 2,863 MT, Portage River is 237
MT, and Sandusky River is 1,100 MT (DAP, Figure D.1, 2023).
Water management practices are pivotal in reducing nutrient loads. These practices, which focus
on detaining or slowing down water flow, settling suspended sediments, and removing dissolved
nutrients from the water column, are instrumental in this process. Two examples are Drainage
Water Management Structures and Conservation Ditches. Another crucial action is the
restoration of wetlands, which intercept and slow runoff, reducing the risk of flooding and
erosion on stream corridors and downstream infrastructure, and improving water quality by
capturing or removing settlement and nutrients through biochemical actions. This water filtering
capability lends some to refer to wetlands as ‘nature's kidneys.’ (DAP, 2023). Nevertheless, we
know that kidneys produce waste products, and we trust that our kidneys will not stop removing
waste products during any part of the year.
Irrespective of the location of a water management project, the potential benefit of nutrient
removal is a universal factor. This benefit is limited by the concentration of the nutrients entering
the wetland structure or generated within the watershed. Coastal wetlands with access to the
western or central Lake Erie Basin face a significant challenge, the 'bathtub effect,' resulting
from water transport and seiche water movement.
The TP and DRP baseline and reduction targets for each watershed are as follows:
23
Table C. Spring value and annual 2008 baseline and reduction targets for DRP and TP by basin.
The 2024 values may be greater than the 2008 baseline due to the construction of additional
animal-feeding operations (DAP, 2023). This underscores the need for recommendations for
evaluating and improving the effectiveness of wetland projects to ensure we are constantly
striving for better solutions.
It is recognized that not all agricultural streams and ditches have nutrient storage and transient
transport characteristics. Particulate P stored on vegetation or the streambed may be remobilized
as particulate P during a high-water discharge event or as SRP if a change in chemical conditions
allows desorption from particles. The instream changes in a watercourse demonstrate that
sediment P associated with iron-oxide (north of the Maumee River) is more readily released
during summer discharges (Field et al., 2023). Therefore, transient storage and release from
headwater instream measures are impacted by both the structures' physical characteristics and the
nutrients' characteristics. The total bioavailability of load (TBAP), along with the bioavailability
of DRP, SRP, and TP (Baker et al., 2019), factors into the efficiency of HAB formation through
adsorption and release (sink and source) of instream structures. The importance of high-water
discharge analysis for effective nutrient reduction cannot be overstated.
OhioEPA acknowledges the changing hydrology of the Maumee River watershed and Ohio, with
more significant flows expected. The US EPA has also recognized these changes and
emphasized the need for the TMDL to reflect the impacts of greenhouse gases and climate
change accurately. This underscores the gravity of the situation and the crucial role in addressing
these challenges, as stated below.
Changes in Watershed Hydrology: As described in Section 4.1.1.7, precipitation, especially in
significant storm events, has increased in the last two decades. These changes have contributed
to a 30 percent increase in DRP loads in the Maumee watershed (Choquette et al., 2019).
Addressing nutrients in the watershed necessarily includes managing the water volume and not
24
just the concentrations of nutrients. Natural infrastructure and controlled drainage have been
identified as cost-effective management practices directed at water management. These practices
help store water on the landscape to infiltrate or be lost through evapotranspiration. With 319 and
GLRI funding, Ohio EPA has collaborated with landowners to install new and emerging water
management technologies, including cascading waterways, water reuse projects (storage and
irrigation), and saturated buffers (OhioEPA Maumee River watershed TMDL, pg. 143, 2023).
The Domestic Action Plan describes the stream processing of nutrients within a wetland or in the
watershed ( DAP 2024). As discussed earlier, the nutrients collect onto suspended sediment.
Once deposited in stream channels, especially in pools, this sediment can be resuspended when
higher stream flows create the necessary forces (DAP 2023, page 68), confirming that the
settling velocity has a significant impact on the sink and source aspect of wetlands (Shapely et
al., 2013). Most of the captured DRP is released back into the stream as the algae die or are
washed off in high flows and via other processes (DAP 2023, page 69) (Withers and Jarvie,
2008). Due to biochemical reactions, DRP can absorb or desorb from streambed sediments. This
generally depends on the nature of the sediment and DRP water concentration (Taylor and
Kunishi, 1971; Kunishi et al.,1972).
Stream bed sediment is known to have a specific phosphorus equilibrium concentration. When
DRP concentrations in water overlying bed sediments are more significant than the sediment's
equilibrium concentration, DRP can be adsorbed. Conversely, DRP can desorb from bed
sediments into overlying waters when the water DRP concentration is below the equilibrium; this
is often described as “internal” loading. During the summers of 2019 and 2021, Kreiling et al.
2022 sampled streambed sediment at 78 sites throughout the Maumee River network (a
significant source of P loads to Lake Erie); micro-relief features and associated vegetation at the
wetland site level influence the rate of nutrient sequestration (Villa et al., 2023).
Williamson's work is significant as it sheds light on the time lag for phosphorus export reduction
after phosphorus watershed imports are abated. This study stresses that the reduction could be
due to the desorption of sediment-bound phosphorus in stream channels, influenced by that
year's reduced DRP ambient water concentrations (DAP 2023 Pages 69-71). The streambed
equilibrium desorption processes are akin to wetland adsorption and desorption, providing a
crucial understanding of phosphorus export reduction.
The study of streambed and sediment analysis focusing on streambed zero P concentration
equilibrium (EPC0) where the SRP concentrations neither sorbs nor desorbs P in the waterway
indicates the stream water SRP concentrations, sediment P saturation, and physiochemical
characteristics influence whether stream sediment function as either a sink or a source (Kreiling
et al., 2023). The sediment of the many miles of the waterways of the Maumee River Basin
functions as a sink and source. The lag time due to the distance from any waterway that functions
as a sink or a source measured at the Waterville measurement station emphasizes the critical
temporal element of when any area functions as a sink or a source of nutrients. The comparison
of the measurement of P in sediment at the waterway to the measurement of P in sediment in the
Maumee Bay (bioavailable P at 12%) and coastal area emphasizes the Stokes Law settling
characteristics of sediment. Also, the estimated 3,000 metric tons removal of P-sediment per year
25
from Maumee Bay and the Maumee River to dry-land areas rather than reintroducing Lake Erie
positively impacts cyanoHAB formation.
The assumption of high phosphorus reductions of up to 80 percent for attainment efficiency
purposes (based upon studies at the Grand Lake St. Mary's) is exceptionally questionable for use
by all wetlands, and it ignores the spring high-water conversion from sink to source in non-water
flow regulated wetlands. The Grand Lake St. Mary’s waterway is a controlled water
environment, and determining nutrient reduction efficiency does not capture the condition of
high discharges, which may be encountered elsewhere in other watersheds. Grand Lake St.
Mary’s complex is an ODNR-engineered wetland complex where incoming water is captured
upstream and pumped through the water control structure, then “released” into riparian or coastal
wetlands designed to provide sediment and nutrient reduction benefits. The water is then
“released through a water control structure,” a diked wetland, or additional processing occurs
before the water reaches the lake. Other wetlands do not contain elaborate water control
mechanisms, and the DRP removal efficiency for each should be calculated individually,
focusing on the zero equilibrium P concentration (EPC0), the soluble reactive phosphorus (SRP)
concentration at which sediment neither sorbs nor desorbs P. They used structural equation
modeling to identify direct and indirect drivers of EPC0.
Stream sediment was a P sink at 40 % and 67 % of sites in 2019 and 2021, respectively. During
both years, spatial variation in EPC0 was shaped by stream water SRP concentrations, sediment P
saturation, and sediment physicochemical characteristics. Agricultural land use and stream size
influenced SRP concentrations and sediment P saturation (PSR) (Kreiling et al., 2022).
The Kreiling analysis is essential when reviewing the nutrient release during the spring critical
loading period between March 1st and July 31st.
Within the Maumee, Portage, and Sandusky watersheds, there have been long-term, large-scale
changes in land management: reduced tillage to minimize erosion and particulate P loss and
increased tile drainage to improve field operations and profitability. These practices can
inadvertently increase liable P fractions (65% SRP load increase) at the soil surface and
transmission of soluble P via subsurface drainage. Findings suggest that changes in agricultural
practices, including some conservation practices designed to reduce erosion and particulate P
transport, may have had unintended, cumulative, and converging impacts contributing to the
increased SRP loads, reaching a critical threshold around 2002 (Jarvie et al., 2017).
Looking at high-water discharges (2-3 times the historical average) from June to July 2015, in
the WLEB, the DRP persisted in soils at or below agronomical levels, signifying legacy-P and
the linkage of sediment-P contribution with cycling and flushing of sediments to cyanoHAB
formation (King et al., 2017).
In Appendix F: Annex 4 Priority Tributary Targets (pages 76 -84) in the 2024 DAP, there is a
detailed review of flows in the numerous tributaries of Lake Erie, comparing nutrient loads to
actual, non-normalized stream discharge. There is general agreement that focusing on spring
discharge is the correct path toward water quality success. Applying the data to mitigation
techniques is necessary to meld science to the reduction of nutrients.
26
Maumee, Portage, Sandusky, and Huron Waterway Discharges
Figure K. Normalized Streamflow Discharge
Figure L. Normalized spring season streamflow discharge and loading (DAP, 2023).
27
Figures M. Spring loading of DRP and TP (OhioEPA Maumee River Watershed Nutrient TMDL, 2023).
Figures K, L, and M demonstrate that attaining the target goal (Table C and Figure L) is far
removed from being achieved. The DRP and TP reductions outlined in Table C will not be
attained without changing the nutrient reduction strategy's basic assumptions and realizing that
wetlands function as sources, especially during the peak water discharge period of March 1st and
July 31st. Without a wholesale change of the assumptions, the 92 Tons of TP reduction from
wetland nonpoint sources for the Maumee River watershed will not be attained, casting a shadow
upon the validity of the 2023 OhioEPA Nutrient Reduction TMDL.
The following points may be ascertained from the ODNR instream water management program
review of whether wetlands are sinks and sources of DRP in forming cyanoHABs.
• In order to determine the theoretical sink or source capabilities of every wetland, a
detailed engineering evaluation should be reviewed for high-water discharge
characteristics (m10%, m10-40%). The ODNR must adhere to the statement, “Because of this
success, ODNR has engineered wetland complexes, and these projects will be designated,
engineered, and constructed at locations that maximize nutrient reduction benefits”
(DAP, 2023). Engineering analysis must be conducted during the high-water discharge
timeframe of March 1st and July 31st.
28
• To determine if existing wetlands are susceptible to high-water discharge (turning a
wetland into a source of nutrients), the same theoretical engineering evaluation should be
undertaken with an increased water flow rate frequency at (m10%, m10-40%) and the
maximum river flow at the tributary in the data collected by King et al.,2022.
• Greater use of the science of high-water flow stream discharges, DRP/TP releases
nutrient cycling, is needed to determine whether wetland and stream buffer APC projects
are meeting the intended attainment goals to prevent ‘fill and spill’ conditions (H2Ohio
Wetland Monitoring Program 2021. H2Ohio Wetland Monitoring Program Plan.
Kinsman-Costello, L.E.).
IV. Conclusions of New and Reconstructed Wetlands Function as Sinks or
Sources in Attaining the TMDL Target During the March 1st through
July 31st Timeframe
The paramount question is when a wetland will function as a source of nutrients for a waterway.
If a wetland functions as a source of nutrients during the high-flow discharge loading period of
March 1st through July 31st, it will be ineffective as a mitigation measure for reducing
cyanoHABs. Wetlands function as both a sink and a source of nutrients during a season, and
determining when the shifts occur is critical to cyanoHAB mitigation planning.
There are many acceptable reasons for constructing wetlands under the ODNR H2Ohio program.
However, some of the wetlands that have been constructed will not function as a sink of nutrients
during the March 1st through July 31st period. A wetland will function as a sink of nutrients
during various conditions, and a wetland will be a source of nutrients during various conditions.
A review of when a wetland functions as a sink and functions as a source of nutrients makes it
highly unlikely that the current mitigation portfolio will meet the 92 metric ton per year
springtime reduction target of TP based upon the 2008 baseline by the target date of 2025
contained in the TMDL. A wetland reduction target has not been established for DRP and N, but
neither has the 92 metric tons per year TP target been adjusted for animal unit growth (leading to
more manure production) in the basin. Likewise, the targets have not been adjusted due to
Climate Change warming trends as recommended by the USEPA.
Fortunately, wetlands function as a control mechanism for N and P. Considering N reductions in
creating more or less toxic HABs in Lake Erie should be factored into attainment planning.
Therefore, a wetland that functions as a sink during the high-flow discharge period will
beneficially abate P and N. Whether that reduction of N will shift the formation of microcystins
to one that has higher toxicity is a question that needs answering.
Measured water discharges to Lake Erie from the Ohio tributaries show between 60 to 98% of
the dissolved reactive phosphorus, total phosphorus, suspended sediment (that contains sediment
-P), and nitrogen-based nutrients are transported during the highest 10%, and the highest 40% of
the tributaries water discharges. The cycling of DRP between the sediment in the Maumee River
tributaries and Lake Erie is a significant source of legacy phosphorus and should be considered
in attainment planning. High-water discharges influence the settling velocity of in-field stream
buffers (e.g., wetlands, flow calming, and riparian restoration), which leads to the DRP adsorbed
by sediment to be released, turning wetlands from sinks to sources of nutrients.
29
High-water discharge and DRP cycling scenarios will negatively influence attaining the 92
metric ton TP/year springtime wetland TMDL reduction target. The failure of the instream
wetland mitigation due to high-water discharge and subsequent failure will influence the number
of cyanoHAB formations and the attainment of the Annex 4 commitment.
Instream-Flow Through, Coastal, Floodplain, and Depressional wetland engineering designs for
P & N nutrients, suspended solid collection, TP, and DRP retention efficiency should be released
for peer review in order to when ascertain whether they perform as sinks or sources, focusing on
the spring loading of March 1st through July 31st. Construction of nutrient-reducing new wetlands
should be paused until a nutrient effectiveness analysis has been completed. Water quality
attainment planning should focus on high-water discharges when most of the cyanoHAB-
forming nutrients are transported and released. Water movement (lag time) between inland
instream wetlands, changing functioning between sources and sinks, and the eventual delivery of
nutrients to western Lake Erie should be part of the wetland evaluation process.
Integrating seasonal Bayesian Changepoint kinetic modeling (incorporating phosphorus and
nitrogen cycles) for evaluating site-specific wetland nutrient reduction effectiveness would be an
effective tool to integrate science into policymaking. Because high-water discharge enormously
impacts the release of TP and DRP, it must become a critical variable in H2Ohio water quality
attainment planning. Flow-weighted mean concentration and flow-weighted mean load, as a tool
to normalize water discharge, should be used with TP, DRP, and N non-normalized load and
concentration to measure progress. High-water discharge is a crucial factor in whether wetlands
function as a sink or a source in producing cycling of bottom sediment (legacy) cyanoHAB
precursors.
References in alphabetical order
Ardon, M., Morse, J.L., Doyle, M.W., Bernhardt, E.S., The water quality consequences of restoring wetland
hydrology to a large agricultural watershed in the southeastern coastal plain. Ecosystems 2010 13 1060-78.
Audet, J., Zak, D., Bidstrup, J., Hoffmann, C.C., Nitrogen and phosphorus retention in Danish restored wet
wetlands. Ambio 49 324-36.
D.B. Baker, L.T. Johnson, R.B. Confesor, et al., Needed: Early-term adjustments for Lake Erie phosphorus target
loads to address western basin cyanobacteria blooms, Journal of Great Lakes Research,
https://doi.org/10.1016/j.jglr.2019.01.011
Baker, B.H., Czarnecki, JMP., Omer, AR., Aldridge, CA., Kroger, R., Prevost, JD., Nutrient and sediment runoff
from agricultural landscapes with varying suites of conservation practices in the Mississippi valley. J. Soil and
Water Conservation. 2018, 73, 75-85
Bhatia, N., Expanding Dredge Geologics, 2016, The Open Workshop.
Berkowitz, J.F., Schlea, D.A., VanZomeren, C.M., Boles C.M.W., Coupling watershed modeling, public
engagement, and soil analysis improves decision making for targeting P retention what land locations. 2020.
Journal of Great Lakes Research 46 (2020) 1331-1339.
Bocaniov, Serghei, Scavia, D., Cappellen, Philippe E., (1). Long-term phosphorus mass-balance of Lake Erie
(Canada-USA) reveal a major contribution of in-lake loading. Ecological Informatics 77 (2023) 102131.
30
Bocaniov, S., Scavia, D., Cappellen, P. E.,(2). A mass-balance approach for predicting lake phosphorus
concentrations as a function of external phosphorus loading: Application to the Lake St. Clair- Lake Erie System
(Canada-USA). Aquatic Ecosystem Health and Management 26(4) 20-30, 2023.
Chaf�in, J. D., Westrick, J.A., Reitz, L.A., Bridgeman, T.B., Microcystin congeners in Lake Erie follow the seasonal
pattern of nitrogen availability. Harmful Algae 127(2023)102466.
Chapra, Steven C., Waveland Press, Chapter 17, Surface Water-Quality Modeling. 1997 reissued 2008. ISBN 1-
57766-605-4.
Cheng F. Y., Basu N. B., Biochemical hotspots: role of small water bodies in landscape nutrient processing. Water
Resources. (2017) 53 5038-56.
Christianson, R., Christianson LE., Wong, C., Helmers, M., McIsaac, G., Mulla, D., McDonald, M., Beyond the
nutrient strategies: common ground to accelerate agricultural water quality improvement in the upper Midwest.
2018. J. Environmental Management 2017. 11.051.
Droste, Ronald and Gehr, Ronald, John Wiley and Sons, Theory and Practice of Water and Wastewater Treatment.
2019, Second Edition. LCCN2018012664.
Field, H. R., Sawyer, A. H., Welch, S. A., Benefiel, R. K., Mathie, D. M., Hood, J. M., et al. (2023). Importance of
dense aquatic vegetation in seasonal phosphate and particle transport in an agricultural headwater stream. Water
Resources Research, 59, e2022WR033782. https://doi.org/10.1029/2022WR033782
Gibbons, K.J., Bridgeman, T., Effect of temperature on phosphorus flux from anoxic western Lake Erie sediments.
2020. Water Resources.
Hull, John. Toledo harbor sediment use plan prepared for the Toledo harbor dredge task force Great Lakes
restoration initiative project. 2012. Hull & Associates.
Jarvie, H.P., Johnson, L. T., Sharpley, A.N., Smith, D.R., Baker, D.B., Bruulsema, T.W., Confesor, R., Increased
soluble Phosphorus Loads to Lake Erie: Unintended Consequences of Conservation Practices? J. Environ. Qual.
46.123-132(2017).
Jiang, B., Mitsch. W J., Influence of hydrologic conditions on nutrient retention, and soil and plant development in a
former central Ohio swamp: a wetlaculture mesocosm experiment Ecological Engineering 2020 157 105969.
Kiekbusch J. J., Schrautzer, J., Nitrogen and phosphorus dynamics of a rewetted shallow flooded peatland. Science
of the Total Environment 2007 380 3-12.
Kamrath, Brock, Yuan, Yongping, Streamflow duration curve to explain nutrient export in Midwestern USA
watershed: Implication for water quality achievements. J. Environmental Management 336 (2023) 117598.
King, Whitney M., Curless, Susan E., Hood, James M., River phosphorus cycling during high flow may constrain
Lake Erie cyanobacteria blooms. Water Research 222 (2022) 118845.
King, Kevin W., Williams, M. R., Johnson, L.T., Smith, D.R., La Barge, G.A., Fausey, N.R., Phosphorus
Availability in Western Lake Erie Drainage Waters: Legacy Evidence across Spatial Scales. J. Environ. Qual.
46:123-132(2017).
Kreiling, Rebecca M., Perner, Patrik M., Breckner, Kenna J., Williamson, Tanja N., Bartsch, Lynn A., Hood James
M., Manning, Nathan F., Johnson, Laura T., Watershed-and reach-scale drivers of phosphorus retention and release
by streambed sediment in a western Lake Erie watershed during summer. Science of The Total Environment 863
(2022) 160804.
31
H2Ohio Wetland Monitoring Program 2021. H2Ohio Wetland Monitoring Program Plan. Kinsman-Costello, L.E.,
Fussell, K., Winslow, C., Kerns, J., Newell, S., Bridgeman, T., Becker, R., Chaffin, J., Doro, K., Johnson, L., Liu,
G., McCluney, K., Michaels, H., Midden, R., Qian, S., Mendonca, R., Wright, N. Lake Erie and Aquatic Research
Network (LEARN) for the Ohio Department of Resources (ODNR). Columbus, OH. 114 pp.
Land, Magnus, Graneli, W., Grimvall, A., Hoffmann C.C., Mitsch, W.J., Tonderski K.S., Verhoeven J. T.A., How
effective are created or restored freshwater wetlands for nitrogen and phosphorus removal? A systematic review.
Environmental Evidence (2016) 5:9.
Miller, SA and Lyon SW, Tile Drainage Increases Total Runoff and Phosphorus Export During Wet Years in the
Western Lake Erie Basin. 2021 Front. Water 3-757106.
Mitsch, W. J., Solving Lake Erie harmful algae blooms by restoring the great black swamp in Ohio. 2017.
Ecological Engineering 108 (2017) 406-413.
Ohio Lake Erie Commission (OLEC). 2024 Ohio’s Domestic Action Plan 2023 to Address Nutrients. Published at:
lakeerie.ohio.gov/planning-and-priorities.
Pace, S., Hood, J M., Raymond, H., Moneymaker, B., Lyon, SW., High-Frequency Monitoring to Estimate Loads
and Identify Nutrient Transport Dynamics in the Little Auglaize River, Ohio. Sustainability, 2022, 14,16848.
Pennuto, C., M.., Burlakova, L.E. , Karatayev, A.Y., Kramer, J., Fischer, A., Mayer, C.,
Spatiotemporal characteristics of nitrogen and phosphorus in the benthos of nearshore Lake Erie
J. Great Lakes Res., 40 (2014), pp. 541-549.
Speir, S.L., Tank, J.L., Mahl, U.H., Quantifying denitrification following floodplain restoration via the two-stage
ditch in an agricultural wetland.2020. Ecol. Eng. 155,105945.
Toledo-Lucas County Port Authority, Dredge material center of innovation AAPA 2020 environmental improvement
awards application.
Case study series, Toledo Harbor Revisited: Changing open water placement policy for western Lake Erie. 2005.
Create Lakes Commission.
Ury, Emily A., Arrumugam, P., Herbert Ellen R., Badiou, P., Page, B., Basu, Nandita, B.B., Source or sink? Meta-
analysis reveals diverging control of phosphorus retention and release in restored and constructed wetlands.
Environmental Research Letters 18 (2023) 083002.
U.S. Army Corps of Engineers Buffalo District, U.S. Army Corps of Engineers dredged material management plan
Toledo Harbor, Ohio.
Villa, Jorge A., Bohrer, G., Ju Y., Wrighton, K., Johnson, N., Kinsman-Costello, L. Carbon sequestration and
nitrogen and phosphorus accumulation 1 in a freshwater, estuarine marsh: effects of microtopography and nutrient
loads. Geoderma Vol 43, Feb 2023,116349.
Wang, W.T., Zhang, T.Q., Zhao, Y.C., Ciborowski, J.J. H., Zhao, Y.M., O’Halloran, I.P., Qi, Z.M., Tan, C.S.,
Characterization of sedimentary phosphorus in Lake Erie and on-site quantification of internal phosphorus loading.
Water Research 188 (2021) 116525.
Withers, PJA., Jarvie, HP., Delivery and cycling of phosphorus in rivers: a review. 2008, Science of the Total
Environment 400: 379-395.
Williamson, T. N., Shaffer, K. H., Runkle, D. L., Hardebeck, M. J., Dobrowolski, E. G., Frey, J. W., Weaver, T. L.
(2021). Nutrient and suspended-sediment concentrations in the Maumee River and tributaries during 2019 rain-
induced fallow conditions. Journal of Great Lakes Research, 47(6). doi:https://doi.org/10.1016/j.jglr.2021.10.004
